Is critical realism a modal realism?

One of Roy Bhaskar’s central ontological claims is that in addition to the actual – the things and events that occur in the material universe – our ontology must also recognise a domain of the real, which includes the actual, but extends beyond it. In his first book, A Realist Theory of Science, he argued that the non-actual includes real causal powers, a very strong argument that I explained towards the end of my last blog post. However, if we accept this argument this opens up rather a large question: what else could be real but not actual? This is a crucial question for materialist realists, because it provides a door through which non-material stuff could reappear in our ontology, and we only want to open that door rather carefully! This post suggests one way of thinking about what sorts of stuff should be admitted.

I can’t offer a carefully justified set of logical criteria in response to this question (one reason why this is a blog post, and rather a speculative one at that, instead of a journal article). But I have often toyed with the idea that what we should admit into the real, along with actual things and events, are the members of certain modal classes (though I am probably using this term in a way that most philosophers wouldn’t recognise). Modality is rather a loose concept, and although I have a fairly precise version of it in mind, I confess it is difficult to express it clearly, so I will work towards it gradually here. Let me start by saying that modalities in the sense I have in  mind are types of relation to actual existence. Existence itself is one of these (the relation being that is IS actual existence), so actuality is a modality, and actual things are members of a modal class that we call the actual.

Image source: Wikipedia (CC BY 2.0)

Beyond that, perhaps the most obvious candidate is absence, an important element of Bhaskar’s later philosophy (though I’m not endorsing his later philosophy in general). In my last post I suggested that “there are true facts about the world which are not facts about actual things”, and absence is one of these. It is true, for example, that there is not an elephant in my room as I type this (not a literal material one anyway!). That’s a true fact about the world, but it’s not a fact about an actual thing – there isn’t an actual elephant for it to be a fact about. There are many specific or ‘concrete’ absences, and they are members of the modal class ‘absence’.

Real causal powers are members of another modal class which we might call necessary-potentiality. Given the physical nature of the universe, it is necessarily true that certain types of structure will have certain causal powers (where a casual power is a potential to have a certain sort of effect on the world) and this is true whether or not any instances of that type of structure actually exist. Again, a true fact about the world, but not necessarily a fact about any actual thing.

Discussions of modality and its ontology in mainstream philosophy, however, have been clustered for some time around another modal class: possibility. This starts from the belief that we can make true claims about possibility – one that pops up in the literature from time to time is “I could have been a dentist”. I’m not sure how we could say whether a claim like that one was true, but some claims about possibility can indeed be true. Here’s one: if I’d said yesterday “It’s possible I will write a blog post tomorrow”, then that claim must have been true since I actually am writing one now. So this looks like another true fact about the world, but at the time that I made the comment it was not a fact about an event that had actually occurred.

Image Source: Wikipedia (public domain)

How could we accommodate possibility in ontology? The best known answer to that question is David Lewis’s claim that alongside our own world there is an infinite set of other possible worlds, and that all of these worlds exist. As Ted Parent puts it in the entry on ‘Modal Metaphysics‘ in the Internet Encyclopaedia of Philosophy, 

“Lewis makes clear that there is only one kind of being, and that all possibilia (that is, all actual and non-actual possible objects) have it. Thus Lewis provocatively suggests that non-actual possibles exist in just the same way that you and I do”

and here he cites Lewis’s 1986 book On the Plurality of Worlds (pp. 2-3). Thus the possible is admitted into our ontology by treating it as existing, but just in a different world: it is possible I could have been a dentist because there is another world in which I am. These days, if you call yourself a modal realist, philosophers will generally assume that you agree with Lewis, but this isn’t really realism about modality, it’s realism about possible worlds.

I think that Lewis’s argument is ridiculous. Let’s just take one argument against it (Parent discusses many others). Imagine that on Monday I said “it’s possible I will write a blog post tomorrow” but as it turned out, I didn’t do it. Lewis’s argument, as I understand it, would be that I was right to say it was possible because there is another world in which I did write the blog post on Tuesday. But when I said that I might write a blog post, I didn’t mean the Dave Elder-Vass in some other world, so even if there was a Dave Elder-Vass in another world that wrote a blog post that wouldn’t establish that I was right when I said that this me could write one. For that argument to work, worlds would have to branch off continuously: if that was so then after I said that on Monday our world would have proliferated and split into many different worlds, in some of which I wrote the blog post and in some of which I didn’t. But that conflicts with everything we think we know about physics and the material world, and given that our scientific knowledge is massively better established than Lewis’s speculation, that means we should reject his argument.

I’m afraid I’m not ready to propose a better way of handling possibility, though it seems possible (!) that it would fit into the argument I’ve been making above, in which case there would be a real but non-actual modal class of the possible.

I suggest that this approach to real but non-actual modal classes is a much more plausible variety of modal realism than Lewis’s.It’s also a variety that seems to fit with Bhaskar’s own arguments, at least up to a point, and one good reason for thinking so is that Bhaskar himself briefly endorses a variety of modal realism in his book of interviews with Mervyn Hartwig, The Formation of Critical Realism. Possible alternatives, he argues, are “in” this world,

“enfolded within the things and structures that we have in this world, and so this world could be different. Modal realism is indispensable for concrete utopianism and for human freedom, and even for efficacy of critique, because you have to be able to say I need not believe this, there are other possibilities” 

and he goes on to link this rather sketchily to “the reality of causal powers, liabilities and tendencies” (p. 66). He returns to the point on the next page, arguing

“It is not just that the world could have been different… but that in our world, the world that we have and know, there are alternatives. But the very idea of an alternative is something inconsistent with actualism” (p. 67).

Actualism, for Bhaskar, is the error of failing to acknowledge the non-actual real, including causal powers and absences, so he does seem to be linking modalities to the non-actual real in a way that is at least somewhat similar to the argument I have sketched out above. Still, there are also signs that he uses the term modality rather more widely than I have here. In the article on modality in the Dictionary of Critical Realism, (edited by Mervyn Hartwig but with more than a trace of Bhaskar’s own style evident in many articles) a variety of forms of modality proliferate (p. 316), before we are brought back to the core issue:

“Since for CR the necessary ways of acting of things are only contingently manifested in actual events, the domain of the real is also the domain of possibility. The actual is thus only a tiny part of the possible… CR espouses modal realism, the view that modal facts exist independently of our knowledge of them, but stops short of endorsing the notion of the actual existence of endless parallel universes” (pp. 316-7).

Critical realism, then, is a kind of modal realism, and we have good reasons to include the modal classes discussed in this post in an ontology that remains a materialist ontology with regard to the actual. But I propose that no others should be allowed through the door until equally plausible cases can be made for admitting them.